pre-registered) and interest in the other direction of effects is formally renounced, my experience is that it is often used cynically and inappropriately to "rescue" a marginally significant result, even when the other tests are two-tailed. 9. line 436: " Although the power of a statistical test could simply be increased by one-sided tests, hardly any study uses directed tests" - increasing power is not a good reason to use directed tests; increasing power by changing the alpha criterion should only be done with a strong justification of having a laxer standard for confirmed results.
← all excerpts
Changes in methodological study characteristics in psychology between 2010-2021.
1
—
—